Date of Award
12-8-2018
Document Type
Thesis
Degree Name
Computer Science, MS
First Advisor
Hung-Chi Su
Committee Members
Hai Jiang; Jeff Jenness
Call Number
ISBN 9780438706750
Abstract
Typically, modern Automatic Speech Recognition (ASR) engines are developed using artificial neural networks (ANN) that utilize probabilistic and stochastic models for analyzing data and computing outputs. As such, limitations are seen with this structure of implementation due to the volume of data required for accuracy and scalability. Resultantly, different software solutions are being considered to improve the accuracy of ASR engines in these unique cases where commonly used models fall short. The presented research illustrates the potential of a biological neural network (BNN) if implemented as an ASR engine. Theoretical comparisons are made between the efficiency of a BNN and a finite-state transducer model (FST). Tests are performed using Numenta Platform for Intelligent Computing (NuPIC) and Kaldi. Kaldi is an OpenFst ASR engine and NuPIC is a BNN platform based on hierarchical temporal memory (HTM) that is configurable for various applications.
Rights Management
This work is licensed under a Creative Commons Attribution 4.0 International License.
Recommended Citation
Walters, Logan Matthew, "A Speech Recognition Noise Analysis Using a Biological Neural Network" (2018). Student Theses and Dissertations. 480.
https://arch.astate.edu/all-etd/480
